A lipase catalyzed condensation reaction with a tricyclic diketone: yet another example of biocatalytic promiscuity

Article abstract of DOI:10.1016/j.tetlet.2009.06.108

Novozym 435 (a commercially available immobilized form of Candida antarctica lipase B) was found to catalyze a condensation reaction of 5-hydroxy-endo-tricyclo[5.2.1.0^2,6]deca-4,8-dien-3-one with acetaldehyde (enzymatically produced from vinyl acetate in situ) under low water conditions, in presence of 10% organic co-solvent (N,N-dimethyl formamide or pyridine), to form a bis-adduct. Even though the condensation reaction occurred with pyridine (acting as a base catalyst) in the presence of acetaldehyde and in the absence of enzyme, the reaction was very slow as compared to the enzymatic process. Thus, while the non-enzymatic process took 4 days to achieve 100% conversion; in presence of enzyme it was possible within 4 h.

Full text of DOI:10.1016/j.tetlet.2009.06.108

Tetrahedron Letters 50 (2009) 5190–5193
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
A lipase catalyzed condensation reaction with a tricyclic diketone: yet
another example of biocatalytic promiscuity
*
*
Abir B. Majumder, Namakkal G. Ramesh , Munishwar N. Gupta
Chemistry Department, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110 016, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Novozym 435 (a commercially available immobilized form of Candida antarctica lipase B) was found to
catalyze a condensation reaction of 5-hydroxy-endo-tricyclo[5.2.1.02,6]deca-4,8-dien-3-one with acetal-
dehyde (enzymatically produced from vinyl acetate in situ) under low water conditions, in presence of
10% organic co-solvent (N,N-dimethyl formamide or pyridine), to form a bis-adduct. Even though the con-
densation reaction occurred with pyridine (acting as a base catalyst) in the presence of acetaldehyde and
in the absence of enzyme, the reaction was very slow as compared to the enzymatic process. Thus, while
the non-enzymatic process took 4 days to achieve 100% conversion; in presence of enzyme it was possi-
ble within 4 h.
Received 28 May 2009
Revised 18 June 2009
Accepted 24 June 2009
Available online 27 June 2009
Keywords:
Aldol condensation
Candida antarctica lipase B
Non-aqueous media
Tricyclic 1,3-diketone
Vinyl acetate
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
frequently used commercial lipase (Novozym 435) under low
water conditions.
In 1997, Sheldon stated that the fine chemical industry, in con-
trast to bulk chemicals manufacturing, is reluctant to apply cata-
lytic technologies.1 In the last decade, there have been vigorous
efforts to leverage biotechnological toolbox to evolve green
chemical approaches.2 As Jaeger had put it more picturesquely, en-
zyme-catalyzed processes are slowly replacing ‘fire and sword’
chemistry.3 The key to strengthen this approach is to use usual en-
zymes in different ways by exploiting their biological potential.
The use of nearly anhydrous media created an opportunity to use
hydrolases for organic synthesis.4 Among such applications, lipases
top the list. One of the reasons (perhaps the most important one) is
that lipases show tremendous versatility vis-à-vis the range of
substrates and biotransformations they catalyze. Acylation and
esterification have been carried out by lipases with numerous
substrates.5 Vinyl esters are often preferred as acyl donors as the
resulting vinyl alcohol is rapidly converted to acetaldehyde there-
by making the reaction faster and irreversible. Weber et al.6 and
Hogberg et al.7 have shown that in organic solvents, vinyl acetate
can produce acetaldehyde and can result in hemiacetal esters. In
the present work, we describe for the first time a condensation
reaction of acetaldehyde with a tricyclic diketone catalyzed by a
5-Hydroxy-endo-tricyclo[5.2.1.02,6]deca-4,8-dien-3-one (1 or
ent-1, Scheme 1) is an interesting tricyclic 1,3-diketone which ex-
ists in the enol form exhibiting pseudo meso character and a very
important synthon for the synthesis of large variety of naturally
occurring cyclopentanoids,8 cubane-type polycyclic compounds,9
and other pharmaceutically important compounds.10 The fast tau-
tomerization of the cyclic diketone such as 1, makes it exist as a
racemic mixture of the two enol forms 1 and ent-1 (Scheme 1)
and thus opens up the possibility of a dynamic kinetic resolution.
O
O
O
Ac
H
O
O
2
1
O
or
Novozym 435
Ac
H
O
O
O
O
ent-2
ent-1
* Corresponding authors. Tel.: +91 11 2659 6584; fax: +91 11 2658 1102 (N.G.R.);
tel.: +91 11 2659 1503; fax: +91 11 2658 1073 (M.N.G.).
E-mail addresses: ramesh@chemistry.iitd.ac.in (N.G. Ramesh), munishwar48@
yahoo.co.uk (M.N. Gupta).
Scheme 1. The fast enantiomerization of the cyclic diketone 1 (enol forms 1 and
ent-1) and its aimed desymmetrization via enzymatic transacetylation.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.06.108

A. B. Majumder et al. / Tetrahedron Letters 50 (2009) 5190–5193
5191
Table 1
Toward that aim, Novozym 435 and vinyl acetate were chosen to
enantioselectively transacetylate 1 (Scheme 1). Instead, we ob-
served the novel condensation product between the diketone and
acetaldehyde.
Novozym 435 (Novo 435) catalyzed condensation reaction of 1 (or ent-1, Scheme 1)
with vinyl esters and with acetaldehyde in nearly anhydrous conditions and in
solvent-free media
Entry no. Reactant 2
Enzyme
Co-solvent Time
C (%)* Yield** (%)
(excess)
(10%, w/w) (10%, v/v)
2. Results and discussion
O
1
Novo 435
DMF
4 h
100
0
95
—
An excess of vinyl ester with a co-solvent, DMF or pyridine, in
minimum amount required to dissolve 1 in vinyl ester was em-
ployed to maintain a nearly solvent-free condition and to acceler-
ate desymmetrization as compared to racemization which is a key
requirement to achieve an efficient dynamic kinetic resolution.11
The progress of the reaction was initially monitored by thin
layer chromatography (TLC) which indicated a gradual disappear-
ance of the starting material and the appearance of a new spot.
The reaction reached completion in 4 h, which was further con-
firmed by HPLC analysis (Fig. 1). A control without enzyme failed
to achieve this reaction even after 5 days (Table 1, entries 2 and 4).
Surprisingly, this enzymatic product appeared more polar
(Rf = 0.6 in ethyl acetate, hexane, and methanol = 4:2:1) than the
expected acetyl derivative 2 (or ent-2), Scheme 1, (Rf >0.9 in the
same solvent) of the alcohol which, for this purpose, was synthe-
sized using standard procedure.10 Interestingly, when vinyl
propionate was used instead of vinyl acetate (Table 1, entries 5
and 6) the reaction produced the same product exhibiting identical
Rf value in TLC and the same retention time in HPLC. However, the
only difference was that the reaction with vinyl propionate was
found to proceed three times slower than that with vinyl acetate
and under the same conditions, took 12 h for the completion (Table
O
O
2
—
DMF
5 days
4 h
O
O
3
Novo 435
Pyridine
Pyridine
Pyridine
Pyridine
Pyridine
Pyridine
100
0
97
—
O
O
4
—
5 days
12 h
5 days
4 h
O
O
5
Novo 435
100
0
98
—
O
O
6
—
O
O
7
Novo 435
—
100
94
95
H
O
8
4 days 100
H
*
C corresponds to the conversion value obtained by HPLC.
This corresponds to the product yield after work-up. The variations between
**
three sets of experimental results were within 2%.
1, entries 1 and 5). This indicated that, it was the vinyl moiety and
not the acyl group that is participating in the reaction.
After the purification of the enzymatic product by column chro-
matography the 1H NMR spectral data of the product revealed that
it was not the expected acetyl derivative, 2 (or ent-2). Upon further
analysis (with 13C NMR, DEPT, and high resolution mass spectral
data; details given as Supplementary data) the product was identi-
fied to be the bis-adduct of 1 or ent-1 (3, Scheme 2).
Lipases have been reported to give abnormal reactions with vi-
nyl acetate due to its hydrolysis to acetaldehyde.6,7 To confirm that
acetaldehyde produced was involved in the reaction, the reaction
of 1 with an excess of acetaldehyde in presence of Novozym 435
and pyridine (10% v/v) was tried. This reaction led to the formation
of the same product 3 (Table 1, entry 7). Evidently, here the
O
O
H
Novozym 435
O
1
HO
O
OH
(Excess)
DMF or pyridine
O
O
H
O
(10%, v/v)
º
4 C, 4h
O
3
Figure 1. HPLC analysis of the enzymatic reaction: (a) at the start of the reaction,
time t = 0 h showing only the starting material, the enol 1 (Scheme 2); (b) at time
t = 0.5 h showing 30% conversion; (c) at t = 4 h showing complete disappearance of
the enol 1 to the product 3 (Scheme 2).
ent-1
Scheme 2. The formation of an unusual bis-adduct 3.

5192
A. B. Majumder et al. / Tetrahedron Letters 50 (2009) 5190–5193
reaction did not follow the known abnormal pathway giving hemi-
acetal or acetylated hemiacetal derivative of the alcohol as earlier
reported by Hogberg et al.7 The formation of this bis-adduct can be
explained as follows: enzyme-promoted nucleophilic addition of 1
or ent-1 with acetaldehyde initially affording the mono-adduct
which further undergoes second reaction with the loss of water
molecule to afford the bis-adduct in 95–97% yield. A control exper-
iment in the absence of enzyme was carried out with 1 (or ent-1),
acetaldehyde, and pyridine. Even though the reaction afforded the
bis-adduct 3, it was much slower, requiring 4 days for the comple-
tion of the reaction (Table 1, entry 8). Despite the fact that pyridine,
taken as a co-solvent, could catalyze the condensation reaction it
was preferred due to the ease of work-up. On the other hand, the
reaction in DMF, which unlike pyridine was not found to catalyze
or assist the condensation, showed more clearly the role of enzyme
catalysis. Thus it was clear that Novozym 435 was not only catalyz-
ing the acyl-oxygen bond cleavage of vinyl acetate to form acetal-
dehyde but it also facilitated the formation of 3 from 1 with
acetaldehyde.
All these facts collectively point out to a reaction sequence
(Scheme 3) which involves: enzymatic cleavage of vinyl esters to
produce acetaldehyde even under anhydrous conditions (which
has already been reported by Weber et al.6), condensation between
1 and acetaldehyde gives an initial product in situ followed by for-
mation of the bis-adduct (diketo form), which was found to exist in
its enolic form (confirmed by 1H NMR spectra showing two enolic
protons which could be exchanged by D2O and also by FTIR spectra
which did not show any isolated carbonyl frequency around 1700–
1710 cmÀ1). The enzyme (being an immobilized formulation) and
the excess unreacted vinyl ester could be easily recovered and re-
used. The percentage of yield in all the cases was very high,
95–97%.
densation.13 More recently, Li et al. described the first lipase-cata-
lyzed asymmetric aldol condensation (between acetone and
aromatic aldehydes) in presence of water as an illustration of bio-
catalytic promiscuity by the enzyme.14 The present case, to the
best of our knowledge, is the first example ever reported where
Novozym 435 exhibits the behavior of both lipase and aldolase in
a single reaction under anhydrous conditions, and in nearly sol-
vent-free media resulting in a bis-adduct. Interestingly, the tem-
perature used in the present case (4 °C) is lower than that used
by Li et al. or Svedendahl et al. but it was found to give a fast
conversion.
3. Conclusion
Lipases are known to catalyze acylation of alcohols, a reaction
frequently carried out by vinyl acetate.7 In the present case, the
cyclic diketone (exists as an enol) failed to get acylated. Instead,
the cyclic diketone underwent an aldol-type condensation reaction
with acetaldehyde (acetaldehyde was produced in situ from hydro-
lysis of vinyl acetate). In this respect, Novozyme 435 behaved more
like an aldolase under nearly anhydrous media. This opens up a
new synthetic opportunity using lipases.
Ser 108
O
O
H
H
Asp 187
O
N
The enzyme active site involvement (based on the models as re-
ported earlier)12 is given in Scheme 4. The free CAL B active site
contains a serine residue, Ser 108, (a part of the catalytic triad)
which binds vinyl acetate and liberates acetaldehyde as it com-
monly does in the case of hydrolysis and then it is the imidazolium
nucleus of the histidine residue, His 224, which facilitates the con-
densation reaction of acetaldehyde by making it a better electron
acceptor through hydrogen-bonding interactions (Scheme 4).13,14
Svedendahl et al. reported that a mutant of Candida antarctica
lipase B (Ser105Ala) could catalyze efficiently Michael type con-
H
O
N
O
Gln
Thr
H
His 224
H
O
NH2
O
H
H2N
Thr
Oxyanion pocket
Scheme 4. Condensation of acetaldehyde with 1 (based on the model as reported
earlier by Li et al.).14
O
O
OH
O
O
Novozym 435
O
HO
O
HO
O
H
Novozym 435
O
H
O
O
O
O
OH
H
O
H
O
O
O
H
O
Scheme 3. An unusual reaction sequence which involves: enzymatic deacetylation of vinyl esters to produce acetaldehyde even under anhydrous conditions, condensation
between 1 and acetaldehyde gives an initial product in situ followed by formation of the bis-adduct (diketo form), which tautomerizes in its enolic form.

A. B. Majumder et al. / Tetrahedron Letters 50 (2009) 5190–5193
5193
product 3, centrifuged and the recovered solid mass was dissolved
in the eluent (acetonitrile/water = 87:13) and was run in a Zorbax
C-18 reverse phase column (fitted in Agilent 1100 series) at a flow
rate of 1 ml/min and was monitored by DAD–UV at 245 nm. Alco-
hol and the product appeared at 1.5 min and 1.2 min, respectively.
The positions were confirmed after running pure NMR grade
samples.
4. Experimental
4.1. Synthesis of the enol adduct (1 or ent-1)
5-Hydroxy-endo-tricyclo[5.2.1.02,6]deca-4,8-dien-3-one was syn-
thesized by a Diels–Alder reaction between cyclopenten-1,4-dio-
ne(Aldrich, Germany) and cyclopentadiene (Fluka, USA) as reported
earlier by Ramesh et al. (mp 170–172 °C , after recrystallization from
dichloromethane and ethyl acetate) and De Puy et al.10,15 1H NMR
(300 MHz: CDCl3, Me4Si), d (ppm): 1.67 (1H, d, J = 8.4 Hz), 1.78 (1H,
d, J = 8.4 Hz), 3.1(4H, s), 4.9 (1H, s), 5.9 (2H, s), 8.9 (1H, br s).13C
NMR (CDCl3) d (ppm): 43.54, 48.98, 107.87, 132.82, 202.77 (given
as Supplementary data).
Acknowledgments
The work was supported by ‘Core group grant for applied bioca-
talysis’ of the Department of Science and Technology (DST), Govt.
of India. Novozym 435 (Novozyme A/S, Denmark) was a kind gift
from Dr. J. S. Rao, Novozymes, Bangalore, India. A.B.M. thanks CSIR,
India for providing senior research fellowship. Additional financial
support from the fund obtained from the Department of Biotech-
nology (DBT), Govt. of India is also gratefully acknowledged.
4.2. Enzymatic synthesis of 3, the bis-adduct of 1
In a 1.5 ml reactor alcohol 1 (or ent-1, 100 mg, 0.6 mol) was dis-
solved in N,N-dimethylformamide, or pyridine (100 lL, 0.1% water
by GC, Qualigens fine chemicals, India) and then an excess of vinyl
acetate (Merck, Germany) or vinyl propionate (Aldrich, Germany)
or acetaldehyde (anhydrous, Fluka, Switzerland) was added to
make the volume 1 ml. Novozym 435 (100 mg, 10 mg enzyme pro-
tein)16 was added so that its load was 10% (w/w, enol 1). Reaction
was set at 4 °C with an orbital shaking at 300 rpm. After different
time intervals aliquots were taken and were analyzed by TLC and
HPLC (see later). Each of the reaction set was run in triplicates
and the variations between three sets of experimental results were
within 2%. The liquid chemicals were distilled and dried over acti-
vated 3 Å molecular sieves (Merck, India) for overnight before use.
Supplementary data
1H NMR, 13C NMR, Mass spectral data are available. Supplemen-
tary data associated with this article can be found, in the online
version, at doi:10.1016/j.tetlet.2009.06.108.
References and notes
1. Sheldon, R. A. J. Chem. Tech. Biotechnol. 1997, 68, 381–388.
2. Gupta, M. N.; Raghava, S. Chem. Central J. 2007, 1, 17.
3. Jaeger, K.-E. Curr. Opin. Biotechnol. 2004, 15, 269–271.
4.3. Purification and product characterization
4. Adlercreutz, P. In Applied Biocatalysis; Straathof, A. J. J., Adlercreutz, P., Eds.; CRC
Press, 2000; pp 295–313; Carrea, G.; Riva, S. Angew. Chem., Int. Ed. 2000, 112,
2312–2341; Gupta, M. N. In Methods in Non-aqueous Enzymology; Birkhauser:
Basel, Switzerland, 2000; Partridge, J.; Harper, N.; Moore, B. D.; Halling, P. J. In
Enzymes in Non Aqueous Solvents: Methods and Protocols; Vulfson, E. N., Halling,
P. J., Holland, H. L., Eds.; Humana Press, 2002; pp 97–104; Roy, I.; Gupta, M. N.
Tetrahedron 2003, 59, 5431–5436; Majumder, A. B.; Singh, B.; Dutta, D.;
Sadhukhan, S.; Gupta, M. N. Bioorg. Med. Chem. Lett. 2006, 16, 4041–4044.
5. Santaniello, E.; Reza-Elahi, S.; Ferraboschi, P. In Stereoselective Biocatalysis;
Patel, R. N., Ed.; Marcel Dekker: New York, Basel, 2000; pp 415–460.
6. Weber, H. K.; Weber, H.; Kazlauskas, R. J. Tetrahedron: Asymmetry 1999, 10,
2635–2638.
7. Hogberg, H.-E.; Lindmerk, M.; Isaksson, D.; Sjodin, K.; Franssen, M. C. R.;
Jogejan, H.; Wijnberg, J. B. P. A.; Groot, A. Tetrahedron Lett. 2000, 41, 3193–3196.
8. Zhu, J.; Klunder, A. J. H.; Zwanenburg, B. Tetrahedron 1994, 50, 10597; Zhu, J.;
Wang, J. Y.; Klunder, A. J. H.; Liu, J.-Y.; Zwanenburg, B. Tetrahedron 1995, 51,
5847.
9. Dilling, W. L. In Recent Advances in Selected Aspects of Bishomocubane Chemistry
Carbocyclic Cage Compounds; Osawa, E., Yonemitsu, O., Eds.; V.C.H.: New York,
1992; p 249.
10. Ramesh, N. G.; Bakkeren, F. J. A.; Groot, D.; Passamonti, U.; Klunder, A. J. H.;
Zwanenburg, B. Tetrahedron 2001, 57, 9877–9887.
11. Hacking, M. A. P. J.; Wegman, M. A.; Rops, J.; van Rantwijk, F.; Sheldon, R. A. J.
Mol. Catal. B: Enzym. 1998, 5, 155–157.
12. Raza, S.; Fransson, L.; Hult, K. Protein Science 2001, 10, 329–338.
13. Svendendahl, M.; Hult, K.; Berglund, P. J. Am. Chem. Soc. 2005, 127, 17988–
17989.
The enzyme was easily separated from the medium by filtration
and the crude product was recovered by decantation of the unre-
acted vinyl ester contaminated with DMF (or pyridine) followed
by evaporation under strong vacuum. The crude product was puri-
fied by a 15 Â 2 cm silica gel (100–200 mesh) column using an elu-
ent consisting of ethyl acetate, methanol, and hexane (in 8:2:1 v/v
ratio). The fractions were collected and evaporated in a rotor evap-
orator to get the pure bis-adduct 3 (104 mg, 95–97%) as a colorless
solid which was further recrystallized from dichloromethane and
ethyl acetate. (mp 243–245 °C). mmax (neat)/cmÀ1 2590 br, 1600,
1400, 1310, and 715; 1H NMR (300 MHz: CDCl3 with 2 drops of
DMSO-d6; Me4Si) d (ppm): 1.04 (3H, d, J = 7.5 Hz), 1.53 (2H, d,
J = 8.4 Hz), 1.72 (2H, d, J = 8.5 Hz), 2.98 (4H, m), 3.58 (1H, q,
J = 7.5 Hz), 5.82 (s, 4H, merged with a br s, 2H, exchangeable with
D2O). 13C NMR (CDCl3 with 2 drops of DMSO-d6) d (ppm): 16.13,
18.3, 42.6, 42.65, 46.1, 46.56, 51.03, 122.36, 131.41, 131.72,
196.01, 196.63; m/z 351.1598 (M+ C20H22O4 requires 350.1518),
301.1417, 236.0731, 224.5758 (see Supplementary data).
14. Li, C.; Feng, X.-W.; Wang, N.; Zhou, Y.-J.; Yu, X.-Q. Green Chem. 2008, 10, 616–
618.
4.4. HPLC analysis of the enzymatic reaction
15. De Puy, C. H.; Zaweski, E. F. J. Am. Chem. Soc. 1959, 81, 4920.
16. Hobbs, H. R.; Kondor, B.; Stephenson, P.; Sheldon, R. A.; Thomas, N. R.; Poliakoff,
M. Green Chem. 2006, 8, 816–821.
The aliquots taken in different time intervals were diluted with
hexane (10Â) to precipitate the unreacted 1 (or ent-1) and the